In this laboratory we have developed a set of techniques that randomized controlled studies indicate can substantially reduce the motor deficit of patients with mild to moderately severe chronic strokes. The techniques, termed Constraint-Induced Movement therapy (CI therapy), involve motor restriction of the less-affected upper extremity for consecutive weeks while at the same time intensively training the more-affected arm. In past work all CI therapy patients were given treatment for 6 hr/day. The primary difference between CI therapy and conventional physical therapy is in the duration and intensity of the treatment, and it was assumed that 6 hr/day of treatment was necessary to obtain an optimal treatment outcome. However, in preliminary work there is a strong suggestion that 3 hr/day of intensive training can give as large an effect as 6 hr/day at the previous standard intensity. The purpose of this grant is to evaluate these preliminary results concerning the efficacy of 3 hours of CI therapy in a rigorously designed, randomized, blinded study. The study will contain 8 groups in a pair of 2x2 factorial designs, a separate one for higher and lower functioning patients. This separation is necessitated by the fact that some important parameters of treatment must differ for the 2 types of patients to obtain an optimal treatment effect. Duration of treatment per day (6 hr vs. 3 hr) will be varied independently of intensity of training (standard vs high). This design will permit the effects of duration and intensity of training to be examined separately (main effects) as well as to determine whether the effect of intensity varies with training duration (interaction effect). In the preliminary study, the treatment administered to the 3 hr/day group was more intense than the treatment administered to the 6 hr/day group so that duration (hr/day) and intensity (trials/hr) were confounded. This project is needed to rigorously evaluate which is the most efficacious combination of these two parameters for patients for different levels of severity of impairment. CI therapy is beginning to be employed clinically at an accelerating pace. The main drawback in the clinical use of this intervention is its recommended duration and intensity, which is expensive in terms of costly therapist time. If 3 hr/day of intense treatment gives as good a result as 6 hr of less intense treatment, the cost of the therapy would be substantially reduced making it feasible to provide it to a larger number of patients than is presently possible.